The country of Botswana, Africa, has the highest rates of HIV infection in the world. Moreover, an emerging drug abuse problem in this part of the world may contribute to immune suppression and to the decline of nutritional status, which in turn facilitates HIV disease progression. This application is based on our African studies indicating that multivitamin supplementation significantly increased CD4 count in pregnant women over time, and our US studies showing that selenium is a powerful predictor of HIV mortality and that selenium supplementation slows HIV disease progression. The overall goal of this application is to determine whether supplementation with multivitamins and selenium can improve immune function and prolong time to AIDS or mortality in HIV infected adults in Botswana. We will recruit 828 participants from the Botswana screening clinics managed by the collaborating Harvard AIDS Institute. Participants with CD4 count >350 will be randomized into a factorial design with 4 groups: combination of multivitamins with selenium, or multivitamins alone, or selenium alone, or placebo, for 24 months. The specific aims of this proposal are 1) To investigate whether the combination of multivitamins and selenium is preferable to either multivitamins alone or selenium alone with respect to improving immune function. 2) To investigate whether multivitamin supplementation is preferable to placebo with respect to improving immune function. 3) To investigate whether selenium is preferable to placebo with respect to improving immune function. 4) To assess how drug abuse affects nutritional status and HIV disease progression in HIV infected men and women. This proposed investigation will determine the efficacy of nutrient therapy to improve immune function and slow HIV disease progression in Botswana. Since antiretroviral therapies are not readily available to HIV infected adults in most countries in Africa, this study may provide an alternative approach to retard HIV progression.